Should Physical Activity Programs be Tailored when Older Adults. Have Compromised Function?

Transcription

1 Page 1 of 23 Should Physical Activity Programs be Tailored when Older Adults Have Compromised Function? Anthony P. Marsh, Ph.D., Elizabeth A. Chmelo, M.S., Jeffrey A. Katula, Ph.D., Shannon L. Mihalko, Ph.D., and W. Jack Rejeski, Ph.D. Affiliations: Department of Health and Exercise Science Wake Forest University, NC, USA Corresponding Author: Anthony P. Marsh, PhD Department of Health and Exercise Science Wake Forest University Winston-Salem, NC Phone: (336) Fax: (336) marshap@wfu.edu Running Head: Tailoring Activity in Older Adults Words:

2 Page 2 of 23 Should Physical Activity Programs be Tailored when Older Adults Have Compromised Function? 1

3 Page 3 of 23 ABSTRACT (150 words) The purpose of this study was to determine whether a walking program supplemented by tasks designed to challenge balance and mobility (WALK+) could improve physical function more than a traditional walking program (WALK) in older adults at-risk for mobility disability. 31 community dwelling older adults (M+SD age=76+5 yrs; Short Physical Performance Battery (SPPB) score= ) were randomized to treatment. Both interventions were 18 sessions (1hr, 3x/wk) and progressive in intensity and duration. Physical function was assessed using the SPPB and the 400m-Walk time (400m-W). A subset of participants in the WALK group who had relatively lower baseline function showed only small improvement in their SPPB score following the intervention (0.3±0.5), whereas a subset of participants in the WALK+ group with low baseline function showed substantial improvement in their SPPB score (2.2±0.7). These preliminary data underscore the potential importance of tailoring interventions for older adults based on baseline levels of physical function. 2

4 Page 4 of 23 INTRODUCTION Regular physical activity is important in the disablement process because it improves cardiovascular function, muscle strength, and balance which in turn may delay the onset of function limitations and mobility disability (Lord, Ward, Williams, & Strudwick, 1995; Nelson et al., 2004; Nelson et al., 2007; The LIFE Study Investigators, 2006; U.S.Department of Health and Human Services, 1996). Mobility disability dramatically increases the risk for dependency (Hirvensalo, Rantanen, & Heikkinen, 2000), which is not surprising given the critical role of mobility in activities of daily living (Frank & Patla, 2003). Walking is the most common mode of physical activity in older adults (DiPietro, 2001; Eyler, Brownson, Bacak, & Housemann, 2003), but it has been suggested that traditional walking interventions lack the perceptual and decision making complexity of real-world environments (Frank & Patla, 2003). Frank and Patla (Frank & Patla, 2003) suggest a number of tasks that could be integrated into a walking program to challenge vision, balance, and mobility. In addition, although moderate intensity walking by itself can improve physical function (Buchner et al., 1997; Moore-Harrison, Speer, Johnson, & Cress, 2008), adding tasks that challenge specific functional capacities such as balance and strength may augment improvements in physical function and mobility. For that reason, multimodal exercise programs are recommended in guidelines for physical activity in older adults (Cress et al., 2005; Nelson et al., 2007; Paterson, Jones, & Rice, 2007) and have been used successfully to improve outcomes focused on physical function and mobility in older adults (Baker et al., 2007; Nelson et al., 2004; The LIFE Study Investigators, 2006). A large number of studies conducted in older adults have compared different modes of exercise to a control, but few studies exist that have compared walking to an alternative or novel intervention. For example, Rooks and colleagues (Rooks, Kiel, Parsons, & Hayes, 1997) showed that both resistance training and walking improved tests of static balance and stair climbing speed. This study is notable because both interventions were entirely self-paced in contrast to being highly structured. Li et al. (Li, 3

5 Page 5 of 23 Fisher, & Harmer, 2005) reported that 16 weeks of "cobblestone" mat walking improved functional reach, static balance, chair stands and 50 ft walk speed and reduced blood pressure to a greater extent than conventional walking in older adults aged yrs. Recently, Shigematsu and colleagues (Shigematsu et al., 2008) showed that a novel stepping intervention improved a number of lower extremity functional outcomes compared to a walking intervention. However, a gap exists for studies that have compared a walking program to an exercise intervention that combines both walking and challenging balance and mobility tasks in older adults at-risk for mobility disability. Although the evidence is sparse the data suggest that the functional ability of an older adult will influence their response to exercise training. For example, Gill et al (Gill et al., 2002) reported a significant beneficial effect in preventing functional decline using a home-based exercise program in a group of older adults with moderate frailty, but no effect among those with severe frailty. In contrast, Chandler and colleagues (Chandler, Duncan, Kochersberger, & Studenski, 1998) found that the impact of 10 weeks of resistance exercise on chair rise performance was significant in participants who were more impaired. With respect to falls, a meta-analysis by Robertson et al. (Robertson, Campbell, Gardner, & Devlin, 2002) showed that a subgroup of older adults aged > 80 yrs and who had experienced a fall benefited the most from a program of muscle strengthening and balance retraining. Recently, Faber and colleagues showed that there were differential effects to two interventions, one based on mobility activities and one based on Tai Chi, such that frail older adults did not respond while pre-frail older adults showed significant improvements in physical function (Faber, Bosscher, Chin, & van Wieringen, 2006). Similarly, Luukinen and colleagues (Luukinen et al., 2006) reported that while exercise in very old adults (>85 yrs) appears to delay loss in mobility, this was not true in a subset of their sample with severe movement or ADL dysfunction. Therefore, it is not always the case that older adults with the lowest function have the greatest potential for improvement with exercise training, although this has been observed previously (Baker et al., 2007). Rather, those with very low function may have reached a point where their disability can not be treated effectively with traditional exercise interventions (Gill et al., 2002). More work is 4

6 Page 6 of 23 needed to identify exercise interventions that can improve physical function and mobility in older adults with low function. To emphasize that older adults have unique and varied needs dependent upon their functional status and health, the American College of Sports Medicine has published new guidelines for this segment of the population (Nelson et al., 2007). Adults over the age of 65 are encouraged to engage in moderately intense aerobic exercise for at least 30 minutes a day, five days a week and to supplement this activity with balance exercises to reduce the incidence of falls (Nelson et al., 2007). Therefore, it would seem appropriate to design walking programs that include activities to challenge multiple domains such as mobility and balance, to better mimic the complexity of real world settings that older adults encounter as they move about in their daily lives. Therefore, among older adults at-risk for mobility disability, the purpose of this study was to test the hypothesis that a physical activity program supplemented by tasks designed to challenge mobility and balance (WALK+) would improve physical function to a greater extent than a traditional walking program (WALK). METHODS Participants We recruited participants from a database of older adults interested in clinical trials, who were: (a) yrs of age; (b) not participating in a regular physical activity program (i.e., any moderate or strenuous activity lasting > 30 min/session, 3 days/wk); (c) scored < 10 on the Short Physical Performance Battery (SPPB)(Guralnik et al., 1994); (d) scored > 6 on the Pfeiffer Mental Status Scale (Pfeiffer, 1975); (e) had no progressive or debilitating conditions that would limit participation in exercise; and (f) received their physician s approval. Within this sample, we classified individuals as either low or high function based on their baseline SPPB score; low function was classified as a SPPB score of 3-8 and high function was classified as a SPPB score of One of the goals of this exploratory study was to obtain data to 5

7 Page 7 of 23 estimate the effects of each intervention since there are no data in the literature that match the outcomes and interventions used in the current study. This study was approved by the Institutional Review Board and all participants signed an informed consent prior to their participation. Measures We had a two person assessment team who were blinded to treatment and trained by the primary author on all outcomes. They completed the majority of baseline and follow-up assessments. Occasionally a third individual assisted in collection of the outcomes and she was unblinded. We acknowledge this as a limitation of the study; however, any bias is mitigated by the fact that all outcomes were administered using prepared scripts. Further, both the SPPB and 400 m walk are objective measures of physical function which reduces the potential for bias on the part of the assessor. SPPB: The SPPB (Guralnik et al., 1994) is a standardized measure of physical performance that assesses standing balance, usual gait velocity over a 4-m course, and the time to sit down and rise from a chair 5 times as quickly as possible. Each task is scored on a scale of 0 to 4, with 0 indicating the inability to complete the task and 1-4 indicating the level of performance. The total SPPB score ranging from 0 (lowest function) to 12 (highest function) was calculated by adding the scores for the three components. The SPPB scale predicts institutionalization, hospital admission, mortality, and disability and is widely used to identify older individuals with compromised levels of lower extremity function who are at-risk for mobility disability (Ferrucci et al., 2000; Guralnik et al., 1994; Guralnik, Ferrucci, Simonsick, Salive, & Wallace, 1995; Onder et al., 2002; The LIFE Study Investigators, 2006). Intraclass correlation coefficients for the SPPB range from 0.88 to 0.92 for measures made 1 week apart (Ostir, Volpato, Fried, Chaves, & Guralnik, 2002). 400-m Walk (400-m W): Participants were timed while they walked 10 clockwise laps around a 40-m indoor course as quickly as possible without sitting, using an assistive device (including a cane), or the 6

8 Page 8 of 23 help of another person (Rolland et al., 2004). A script was used to provide identical instructions to all participants. The script was as follows: You will be walking 10 complete laps around the course, about ¼ mile. Please walk as quickly as you can, without running, at a pace you can maintain over the 10 laps. After you complete the 10 laps, I will tell you to stop. On each lap the assessor offered standard encouragements ( Keep up the good work. You are going well. Looking good. Well done. Good job. ) and told the participant the number of laps completed and the number remaining. Participants were allowed to rest in place and were given a maximum of 15 min to complete the 400-m walk. At baseline, participants completed the informed consent, the SPPB, a self-administered questionnaire consisting of demographics, health history and psychosocial questions, and 400-m W. Following baseline testing, participants were randomized using a table of random numbers to either the WALK (N=15) or the WALK+ (N=16) group. Follow-up testing was completed within a week after the completion of the intervention. Participants in each group attended a 1 hr session, three times a week, at an indoor exercise facility. With only a few exceptions the intervention was completed in 6 weeks; however, in the event of a missed session (e.g., due to illness, vacation, or schedule conflict), a make-up session was scheduled so that all participants completed 18 sessions. Both interventions exercised in small groups at separate times of the day. Therefore, the WALK group did not see or interact with the members of the WALK+ group nor did they observe any of the equipment used in the WALK+ intervention. The WALK and WALK+ groups were supervised at all times by a lead interventionist with ACSM Exercise Specialist certification. She was assisted by a graduate student and several undergraduate students with training in exercise science. The interventionists did not discuss the other intervention with participants and participants were instructed to continue with their normal routine outside of the intervention sessions. 7

9 Page 9 of 23 Both interventions were progressive in both intensity and duration. Due to the varied functional abilities of our sample, a standardized progression for the walking program was not feasible. For example, at the start of the intervention not all participants could walk for 25 min either continuously or with rest stops and not all participants could walk at or near a Rating of Perceived Exertion (RPE) of Therefore, in the first 1-2 sessions we assessed each participant's capabilities and then encouraged them to increase the duration and/or intensity over the course of the intervention. The goal was to reach or approach 25 min of continuous walking in the case of the WALK group or continuous walking interspersed with the "plus" component for the WALK+ intervention, and to walk at an intensity comparable to an RPE of The interventionists played a key role in tracking each participant's previous efforts and setting a goal for each subsequent exercise session. WALK Intervention Each participant walked two laps at a low intensity before walking for up to 25 minutes at a moderate intensity walking pace, i.e., Rating of Perceived Exertion (RPE) of on the Borg RPE scale (Borg, 1973). Heart rate, blood pressure, and RPE were recorded at the mid-exercise time point. After the walking bout, participants completed two laps at a slow walking pace and 20 minutes of stretching and flexibility exercises for major muscle groups. WALK+ Intervention Participants followed a similar walking protocol as described above, however, they were instructed to complete four obstacle stations as they encountered them along the track. The total time of the walk/obstacle session was a maximum of 25 min of which 8-10 min was devoted to the obstacle stations. After the walk/obstacle phase, participants completed two laps at a slower pace and 10 min of flexibility exercises. At the end of every week, an investigator (EC) evaluated each participant to determine mastery over a progression. Each participant was required to complete a minimum of three sessions at each progression. The same investigator (EC) observed each participant over two sessions. If participants 8

10 Page 10 of 23 could complete the progression with no apparent difficulty, assessed by speed and stability of movement, the investigator allowed them to progress to the next level. If the participant did not feel they could confidently progress to the next level they were not required to do so. Balance. Progression 1 (P1): standing on a square foam pad (AIREX, M-F Athletic, Cranston, RI) for 10 seconds with feet shoulder width apart. The participants were given the option of standing on one foot on the foam pad once they had mastered the first progression. This task was completed before each progression. Then the participants walked for 3 meters between two lines that were approximately sixinches apart. Progression 2 (P2): walking heel to toe on a line at a self-selected pace for a distance of 3 meters. Progression 3 (P3): walking on a 3.25 m long foam balance beam (AIREX, M-F Athletic, Cranston, RI). Progression 4 (P4): walking heel to toe on the foam balance beam. Progression 5 (P5): walking with each foot on a separate foam balance beam approximately 0.5 m apart. Hurdles. P1: walking over five white lines ~1 m apart drawn on the track without stepping on the lines. P2: walking over six, six-inch hurdles placed ~1 m apart (db Manufacturing, Middleton, WI). P3: do P2 plus walk over a six-inch hurdle, duck under a 4-foot hurdle, step over another six-inch hurdle, and duck under another 4-foot hurdle. P4: same as P2 but walking laterally. P5: laterally traversing over the P3 obstacles while alternating directions, i.e., moving laterally step over hurdle, turn 180 degrees, duck under high hurdle, turn 180 degrees, step over hurdle, turn 180 degrees, and duck under the final high hurdle. Cones. P1: walking over a crooked stick obstacle (db Manufacturing, Middleton, WI) placing one foot on each side of the stick, alternating feet and avoiding stepping on the stick. P2: zigzagging around six cones that were ~1 m apart. P3: walking through the P2 course carrying a dumbbell weight in each hand. P4: walking down and back, over the crooked stick crossing feet over one another (i.e., grapevine maneuver). P5: do P4 while carrying a dumbbell in each hand. 9

11 Page 11 of 23 Walking Trails A and B. This task was a modification of a task described by Alexander and colleagues (Alexander, Ashton-Miller, Giordani, Guire, & Schultz, 2005). In Trails A participants stepped on numerically ordered rubberized circular discs, numbered 1 to 25. The numbers were spread out, in order, about a step apart within a 2 meter wide by 4 meter long rectangle such that participants walked 12 circles down and 13 circles back. In Trails B participants walked from a number to a letter (1-A-2-B-3-C-4- - L-13). Data Analysis Statistical analyses were performed using SPSS v.14.0 (SPSS Incorporated, Chicago, IL). Analysis of Covariance (ANCOVA) was used to examine group differences in the SPPB and 400m-W change scores with the pretest scores being used as covariates in each analysis. To examine whether baseline levels of function moderated the treatment effect in these analyses, we constructed an interaction term by multiplying the baseline scores for function by group assignment. Due to the reduced power of testing interactions in linear models and the exploratory nature of this research, the alpha level for both interaction terms was set at the p <.10 level. We explored interpretation of significant interaction terms by creating a dichotomous variable for baseline function scores within our sample and crossing this variable with group assignment. The dichotomous variable for lower extremity function was created by assigning a 1 (low function) to individuals with a baseline SPPB score of 3-8 and 2 (high function) to individuals with a baseline SPPB score of RESULTS Ten male and 21 female community-dwelling older adults were recruited for this study. The mean (+ SD) age of the participants was yr (range 67-85) with a mean body mass index (BMI) of kg m -2 (range 19-39) and SPPB score of (range 3-10), respectively. There were no significant differences between the WALK and WALK+ group in any demographic or health measure at baseline 10

12 Page 12 of 23 (Table 1). One participant from the WALK group and one participant from the WALK+ group dropped out of the study for reasons unrelated to the intervention (previously undocumented dementia, preexisting brain aneurism). Therefore, 29 individuals completed the interventions, 14 in the WALK group and 15 in the WALK+ group. There were no exercise- or testing-related adverse events. There were no significant differences between the WALK and WALK+ group in any outcome measure at baseline (Table 1). In the ANCOVA examining change scores in SPPB, there was a significant group main effect, F(1, 25) = 4.73, p = Paired t-tests (2 tailed) conducted within each group on the baseline and follow-up SPPB score showed that both groups increased their SPPB score following the intervention but only the WALK+ group difference was statistically significant (WALK: 8.1±1.4 to 8.8±2.0, t(13)=-1.93, p=0.075; WALK+: 8.7±1.8 to 10.0±1.4, t(14)=-3.57, p=0.003). Interpretation of this main effect is qualified by a significant interaction term, F (1, 25) = 3.59, p = Recall that to accommodate interpretation of significant interaction terms, we created a dichotomous variable for function, relative to our sample, (low vs. high) and crossed this with the treatment variable (WALK vs. WALK+); these cell means can be found in Table 2. Follow-up tests revealed that this interaction was due to differences in the SPPB change scores between individuals with low baseline function assigned to either the WALK or WALK+ intervention, F(1,12) = 5.38, p = The estimated means suggest that participants assigned to the WALK group with low baseline function showed only small improvement in their SPPB score following the intervention, whereas participants assigned to the WALK+ group with low baseline function showed substantial improvement in their SPPB score. There was no significant difference between the WALK and WALK+ groups in change scores for the 400m-W data, F (1, 25) = 2.11, p = To calculate estimates of effect sizes (ES) for the SPPB and 400 m walk, we used the estimated marginal means (i.e., least squares means), obtained from an ANCOVA on change scores that controlled for baseline score. For the SPPB change scores, the least squares means were 0.6 for WALK and 1.4 for 11

13 Page 13 of 23 WALK+, indicating improvement in lower extremity function, with a common standard deviation of 1.3. The ES was 0.6 which is moderate. For the 400 m walk change scores, the least squares means were seconds for WALK and seconds for WALK+, indicating a faster walk time, with a common standard deviation of 26.4 sec. The ES was 0.2 which is low. Given the exploratory nature of this study we also examined changes in the three components of the SPPB using an ANCOVA that controlled for baseline differences between the groups and used the change score as the outcome. Recall that each component of the SPPB is scored 0-4. There was a significant difference between the WALK and WALK+ groups in balance change score [raw mean±sd for WALK and WALK+; 0.1±0.6 vs. 0.3±0.9, F(1, 26) = 6.02, p = 0.021]. There were no significant group differences for change scores for gait speed (0.4±0.6 vs. 0.4±0.5) or chair rises (0.3±1.0 vs. 0.6±1.0) but the trends in the means indicate improvement in both of these components for both groups. DISCUSSION This study compared the effects of a walking program that integrated complex tasks designed to challenge mobility and balance (WALK+) to a traditional walking program (WALK) on physical functioning in older adults with compromised lower extremity function. There was evidence from the SPPB data that participants with lower physical function at baseline benefited more from the WALK+ program than the WALK program. This was not true for older adults with higher levels of baseline function; although the results did not exceed conventional levels of statistical significance, there was a trend for the higher functioning older adults to benefit more from WALK than WALK+. While these findings are limited by the small number of participants in this study which reduced the power to detect differences between groups, they do suggest that tailoring of treatment may be particularly important in older adult populations. This finding is consistent with the recent guideline published by Nelson et al. (Nelson et al., 12

14 Page 14 of ). In particular, older adults with low function may need to improve balance and mobility skills first to subsequently reap the benefits of a standard walking program. Although several studies suggest that frail older adults may derive limited or no benefit from an exercise program (Faber et al., 2006; Gill et al., 2002; Luukinen et al., 2006), the older adults with very low function did benefit from a program of walking combined with balance and mobility tasks. It is important to acknowledge that the program we implemented might not be applicable to older adults who can not walk short distances or who have SPPB scores of 0-2. In particular, very frail older adults might be at increased risk for falls in the absence of adequate intervention staff supervision which may, in some cases, involve one-on-one interaction. This is likely a reality of any intervention that focuses on mobility and balance in very frail individuals. The concept of tailoring an intervention to the needs of an individual or a particular subgroup is not new (Brawley, Rejeski, & King, 2003; Heath & Stuart, 2002; King, Carl, Birkel, & Haskell, 1988; Singh, 2002). The stepped approach of the levels at three of the four stations and the individualized prescription of the walking program that we used in this study are consistent with this practice. The data suggest that the balance and mobility challenges offered by the four stations were important to the improvement of the sub-sample of low functioning older adults. It is worth mentioning that while two thirds of the 15 participants reached the fifth (highest) level of the hurdle and cones stations, only 1 individual reached the 5 th progression of the balance station by the end of 18 sessions. Six individuals did not progress beyond the third progression at the balance station, while two individuals only made the third progression on the hurdle and cone stations. Anecdotally, the Walking Trails A/B station was very taxing for the majority of participants who appeared to have difficulty maintaining stability while planning their next movement. The trends in the 400m-W data are interesting because less time was spent walking in the WALK+ intervention since participants had to integrate the four stations within the fixed 25 min time period of the 13

15 Page 15 of 23 intervention. Overall, the within group trends observed suggest that the walking component of both the WALK and WALK+ interventions was effective. However, the effect size related to the difference between the groups in 400-m W change scores was low (0.2). The low functioning participants in the WALK+ intervention did appear to benefit in spite of the reduced time spent on the cardiovascular component of the intervention. Preserving the ability to walk 400 meters is important for a wide variety of daily and social activities that only require walking this modest distance (Hadley, 2007). Losing the ability to walk 400 meters forces an older adult to either resort to compensatory strategies that may pose social or financial burdens, or to restrict the scope of their daily lives. Relative to changes in SPPB, the lower functioning participants in the WALK+ condition experienced the most change (M = 2.2 points), whereas the lower functioning participants in the WALK condition experienced the least amount of change (M = 0.3 points). Perera and colleagues (Perera, Mody, Woodman, & Studenski, 2006) examined the magnitude of meaningful change in several measures of lower extremity function and concluded that a 0.5 point change in SPPB score is a small meaningful change and a 1 point change can be considered substantial. The changes we observed in the SPPB are encouraging given the short duration of the intervention. Overall, the effect size related to the difference between the groups in SPPB change scores was moderate (0.6). The analyses of the three components of the SPPB are consistent with the fact that the stations included in the WALK+ intervention place a premium on an individual's ability to maintain stability. The data suggest that the balance component of the SPPB improved more in the WALK+ group compared to the WALK group. The improvement in balance may also be responsible for the trend observed in the chair stands. It is important to recognize that all three components of the SPPB showed trends for improvement following the interventions. An interesting question is whether or not a similar result would have occurred by simply completing the four stations without any walking component. 14

16 Page 16 of 23 The changes observed in the SPPB are important because this outcome is predictive of disability, mortality, and institutionalization (Guralnik et al., 1994). Our data show that the SPPB may be useful in discriminating between older adults in need of tailored interventions and those needing only to increase their level of physical activity. This may be useful for physicians in clinical decision making about physical activity for older patients. Also, it is encouraging that the WALK+ intervention appeared to offer measurable advantages over a standard walking program, a comparison that is more rigorous than one using a non-exercise control. Finally, the results and conclusions of this study should be viewed in the context of several limitations. The primary limitation of this exploratory study was the small sample size which reduced the power to detect differences between the groups. Second, we did not stratify the randomization by baseline function. Finally, as discussed above, an unblinded investigator assisted with some outcome assessments which introduced the potential for bias. Additional research is needed to replicate these findings using larger samples, to study long-term interventions, and to determine whether the benefits demonstrated here translate to other forms of functioning such as the performance of activities of daily living. 15

17 Page 17 of 23 REFERENCES Alexander, N. B., Ashton-Miller, J. A., Giordani, B., Guire, K., & Schultz, A. B. (2005). Age differences in timed accurate stepping with increasing cognitive and visual demand: a walking trail making test. Journals of Gerontology, 60, Baker, M. K., Kennedy, D. J., Bohle, P. L., Campbell, D. S., Knapman, L., Grady, J. et al. (2007). Efficacy and feasibility of a novel tri-modal robust exercise prescription in a retirement community: a randomized, controlled trial. Journal of the American Geriatrics Society, 55, Borg, G. A. (1973). Perceived exertion: a note on "history" and methods. Medicine and Science in Sports, 5, Brawley, L. R., Rejeski, W. J., & King, A. C. (2003). Promoting physical activity for older adults: the challenges for changing behavior. American Journal of Preventive Medicine, 25, Buchner, D. M., Cress, M. E., de Lateur, B. J., Esselman, P. C., Margherita, A. J., Price, R. et al. (1997). A comparison of the effects of three types of endurance training on balance and other fall risk factors in older adults. Aging (Milan, Italy), 9, Chandler, J. M., Duncan, P. W., Kochersberger, G., & Studenski, S. (1998). Is lower extremity strength gain associated with improvement in physical performance and disability in frail, communitydwelling elders? Archives of Physical Medicine and Rehabilitation, 79, Cress, M. E., Buchner, D. M., Prohaska, T., Rimmer, J., Brown, M., Macera, C. et al. (2005). Best practices for physical activity programs and behavior counseling in older adult populations. Journal of Aging and Physical Activity, 13,

18 Page 18 of 23 DiPietro, L. (2001). Physical activity in aging: changes in patterns and their relationship to health and function. Journals of Gerontology, 56 (2), Eyler, A. A., Brownson, R. C., Bacak, S. J., & Housemann, R. A. (2003). The epidemiology of walking for physical activity in the United States. Medicine and Science in Sports and Exercise, 35, Faber, M. J., Bosscher, R. J., Chin, A. P. M., & van Wieringen, P. C. (2006). Effects of exercise programs on falls and mobility in frail and pre-frail older adults: A multicenter randomized controlled trial. Archives of Physical Medicine and Rehabilitation, 87, Ferrucci, L., Penninx, B. W., Leveille, S. G., Corti, M. C., Pahor, M., Wallace, R. et al. (2000). Characteristics of nondisabled older persons who perform poorly in objective tests of lower extremity function. Journal of the American Geriatrics Society, 48, Frank, J. S. & Patla, A. E. (2003). Balance and mobility challenges in older adults: implications for preserving community mobility. American Journal of Preventive Medicine, 25, Gill, T. M., Baker, D. I., Gottschalk, M., Peduzzi, P. N., Allore, H., & Byers, A. (2002). A program to prevent functional decline in physically frail, elderly persons who live at home. The New England Journal of Medicine, 347, Guralnik, J. M., Ferrucci, L., Simonsick, E. M., Salive, M. E., & Wallace, R. B. (1995). Lower-extremity function in persons over the age of 70 years as a predictor of subsequent disability [see comments]. The New England Journal of Medicine, 332, Guralnik, J. M., Simonsick, E. M., Ferrucci, L., Glynn, R. J., Berkman, L. F., Blazer, D. G. et al. (1994). A short physical performance battery assessing lower extremity function: association with self- 17

19 Page 19 of 23 reported disability and prediction of mortality and nursing home admission. Journals of Gerontology, 49, M85-M94. Hadley, E. C. (2007). Testing interventions to preserve walking ability: progress against disability, one step at a time. Journals of Gerontology, 62, Heath, J. M. & Stuart, M. R. (2002). Prescribing exercise for frail elders. The Journal of the American Board of Family Practice, 15, Hirvensalo, M., Rantanen, T., & Heikkinen, E. (2000). Mobility difficulties and physical activity as predictors of mortality and loss of independence in the community-living older population. Journal of the American Geriatrics Society, 48, King, A. C., Carl, F., Birkel, L., & Haskell, W. L. (1988). Increasing exercise among blue-collar employees: the tailoring of worksite programs to meet specific needs. Preventive Medicine, 17, Li, F., Fisher, K. J., & Harmer, P. (2005). Improving physical function and blood pressure in older adults through cobblestone mat walking: a randomized trial. Journal of the American Geriatrics Society, 53, Lord, S. R., Ward, J. A., Williams, P., & Strudwick, M. (1995). The effect of a 12-month exercise trial on balance, strength, and falls in older women: a randomized controlled trial. Journal of the American Geriatrics Society, 43, Luukinen, H., Lehtola, S., Jokelainen, J., Vaananen-Sainio, R., Lotvonen, S., & Koistinen, P. (2006). Prevention of disability by exercise among the elderly: a population-based, randomized, controlled trial. Scandinavian Journal of Primary Health Care, 24,

20 Page 20 of 23 Moore-Harrison, T. L., Speer, E. M., Johnson, F. T., & Cress, M. E. (2008). The effects of aerobic training and nutrition education on functional performance in low socioeconomic older adults. Journal of Geriatric Physical Therapy, 31, Nelson, M. E., Layne, J. E., Bernstein, M. J., Nuernberger, A., Castaneda, C., Kaliton, D. et al. (2004). The effects of multidimensional home-based exercise on functional performance in elderly people. Journals of Gerontology, 59, Nelson, M. E., Rejeski, W. J., Blair, S. N., Duncan, P. W., Judge, J. O., King, A. C. et al. (2007). Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Medicine and Science in Sports and Exercise, 39, Onder, G., Penninx, B. W., Lapuerta, P., Fried, L. P., Ostir, G. V., Guralnik, J. M. et al. (2002). Change in physical performance over time in older women: the Women's Health and Aging Study. Journals of Gerontology, 57, M289-M293. Ostir, G. V., Volpato, S., Fried, L. P., Chaves, P., & Guralnik, J. M. (2002). Reliability and sensitivity to change assessed for a summary measure of lower body function: results from the Women's Health and Aging Study. Journal of Clinical Epidemiology, 55, Paterson, D. H., Jones, G. R., & Rice, C. L. (2007). Ageing and physical activity: evidence to develop exercise recommendations for older adults. Canadian Journal of Public Health, 98 Suppl 2:S , S Perera, S., Mody, S. H., Woodman, R. C., & Studenski, S. A. (2006). Meaningful change and responsiveness in common physical performance measures in older adults. Journal of the American Geriatrics Society, 54,

21 Page 21 of 23 Pfeiffer, E. (1975). A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients. Journal of the American Geriatrics Society, 23, Robertson, M. C., Campbell, A. J., Gardner, M. M., & Devlin, N. (2002). Preventing injuries in older people by preventing falls: a meta-analysis of individual-level data. Journal of the American Geriatrics Society, 50, Rolland, Y. M., Cesari, M., Miller, M. E., Penninx, B. W., Atkinson, H. H., & Pahor, M. (2004). Reliability of the 400-m usual-pace walk test as an assessment of mobility limitation in older adults. Journal of the American Geriatrics Society, 52, Rooks, D. S., Kiel, D. P., Parsons, C., & Hayes, W. C. (1997). Self-paced resistance training and walking exercise in community-dwelling older adults: effects on neuromotor performance. Journals of Gerontology, 52, M161-M168. Shigematsu, R., Okura, T., Nakagaichi, M., Tanaka, K., Sakai, T., Kitazumi, S. et al. (2008). Squarestepping exercise and fall risk factors in older adults: a single-blind, randomized controlled trial. Journals of Gerontology, 63, Singh, M. A. (2002). Exercise to prevent and treat functional disability. Clinics in Geriatric Medicine, 18, The LIFE Study Investigators (2006). Effects of a Physical Activity Intervention on Measures of Physical Performance: Results of the Lifestyle Interventions and Independence for Elders Pilot (LIFE-P) Study. Journals of Gerontology, 61, U.S.Department of Health and Human Services (1996). Physical activity and health: A report from the Surgeon General Atlanta: U.S. Department of Health and Human Services, Center for Disease Control and Prevention, GA, National Center for Disease Prevention and Health and Promotion. 20

22 Page 22 of 23 Table 1: Baseline characteristics including physical function outcome measures of the participants. Data are mean (SD) unless otherwise noted. There were no significant differences between the groups at baseline. Variable WALK WALK+ Total Number of Participants Age (yrs) 77.3 (5.1) 74.4 (5.2) 75.8 (5.3) # of Males/Females 5/10 5/11 10/21 Height (cm) (10.3) (9.2) (9.6) Body mass (kg) 78.1 (14.1) 74.1 (17.7) 76.0 (15.9) BMI (kg/m 2 ) 29.9 (4.6) 27.9 (6.1) 28.7 (5.0) Ethnicity N (%): Caucasian 14 (93) 12 (75) 26 (84%) African American 1 (7) 4 (25) 5 (16%) Education N (%): < High School 7 (47) 6 (38) 13 (42%) > College 7 (47)) 7 (44) 14 (45%) Marital Status N (%): Married 9 (60) 8 (50) 17 (55%) Widowed 4 (27) 6 (38) 10 (33%) Divorced 2 (13) 1 (6) 3 (9%) Never Married 0 1 (6) 1 (3%) Living Alone N (%) 4 (27) 5 (31) 9 (29%) Comorbidities N (%): Diabetes 3 (20) 3 (19) 6 (19%) Hypertension 10 (67) 13 (81) 23 (74%) COPD 2 (13) 2 (13) 4 (13%) Arthritis 11 (73) 11 (69) 22 (71%) SPPB (0-12) m-W (s) COPD: Chronic Obstructive Pulmonary Disease SPPB: Short Physical Performance Battery 400m-W: Time to walk 400 m

23 Page 23 of 23 Table 2: Estimated marginal means and standard errors for change in Short Physical Performance Battery (SPPB) and 400 m walk time (400m-W) for participants categorized as either low or high function within the sample assigned to the WALK and WALK+ groups. Baseline Group Change in Change in Function SPPB 1 400m-W (s) Low* High WALK n=10 WALK+ n=5 WALK n=4 WALK+ n= (.5) -9.6 (8.5) 2.2 (.7) (11.5) 1.4 (.7) (13.3) 1.0 (.5) (8.2) 1. Baseline function by Group interaction was significant (p = 0.046). * Significant difference between WALK and WALK+ for Low in change in SPPB (p = 0.039)